LIMNOLOGY
May 1991
AND
OCEANOGRAPHY
Oceanogr., 36(3), 199 1, 4 13-423
Limnol.
0 1991, by the American
Society of Limnology
and Oceanography,
Volume
36
Number
3
Inc
The major and minor element geochemistry of Lake Baikal
Kelly Kenison Falkner, ’ Chris I. Measures, 2 Sarah E. Herbelin, 3 and John A4. Edmond
Earth, Atmospheric and Planetary Sciences, Massachusetts Institute of Technology, Cambridge 02 139
Ray F. Weiss
Scripps Institution of Oceanography, UCSD, La Jolla, California 92093
Abstract
A comprehensive, joint Soviet-American study of the chemistry of Lake Baikal, the world’s
deepest (1,632 m) lake, was carried out in July 1988. In this paper, we report the major, minor,
and preliminary trace element concentrations for three profiles obtained at or near the deepest and
central part of the three major basins of the lake. With the exception of Ba, the distributions of
major and minor elements were homogeneous, displaying no variations greater than analytical
uncertainties. Average concentrations in pmol kg- I (1 SD) are titration alkalinity = 1,093(6), SOd2= 57.4(1.3), Cl = 12.3(0.7), Ca = 402(7), Mg = 126(l), Na = 155(4), and K = 24.1( 1.O); and in
nmol kg-’ are Sr = 1,350(30), Li = 296(12), Ba = 74.7(2.6), Rb = 7.10(0.23), and U = 1.77(0.12).
Excluding K and Cl, these values compare favorably with previously published results. Although
some hydrothermal activity is known to occur within the lake, it does not appear to significantly
affect major ion cycling. The residence times of the major ions are 330 yr or the same as that of
water in the basin and so are controlled predominantly by their riverine fluxes. There is not yet
enough information to assess whether hydrothermal processes affect minor element cycles. Ba
concentrations decrease with depth, showing abrupt decreases near the bottom at two stations. It
appears to undergo some form of uptake at the sediments, but further study is required to discern
the processes governing Ba distribution.
Lake Baikal, located in an intracontinental rift zone in the central region of southern
I Present address: Centre National D’Etudes Spatiales, GRGS, 18 Ave. Edouard Belin, 3 1055 Toulouse
Cedex, France.
2 Present address: Department of Oceanography,
University of Hawaii, Honolulu 96822.
3 Present address: Chemistry Department, Reed College, Portland, Oregon 97202.
Acknowledgments
We thank T. Bowers for assistance with EQ3nr, I.
Ellis and E. Boyle for Cu, Ni, Cd, and Zn analyses, R.
Mortlock and P. Froelich for Ge analyses, and M.
Palmer for the ID-TIMS and Sr isotope measurements.
We are grateful to R. Williams, E. Carmack, C. Lange,
and P. Salameh for assistance with the hydrographic
work, sample collection, and data processing, and to
the officers and crew of the RV Vereshchagin for help
with all aspects of the fieldwork. Logistical support in
Siberia, is the world’s deepest (1,632 m,
Weiss et al. 199 1) and most voluminous
lake (23,000 km3, Kozhov 1963). The Baikal tectonic depression is believed to have
formed as early as the Paleozoic with the
three lacustrine basins merging to their
present configuration
in the late Tertiary
(-60 Ma, Kozhov 1963). Having the lonthe Soviet Union was provided by Goskomgidronet
through the assistance of V. Koropalov and his staff
from the Institute of Applied Geophysics in Moscow,
and by the Siberian Branch of the USSR Academy of
Sciences, with the help of M. Grachev and Y. Kusner
of the Limnological Institute at Irkutsk.
Funding for U.S. participation in the fieldwork and
for sample collection was provided by the U.S. National Science Foundation. Funding for the analytical
work was obtained by various means.
413
414
Falkner et al.
WN
54’N
Fig. 1. Map of stations occupied during the Lake
Baikal expedition. Water samples for major ions and
trace elements were obtained at stations 7, 17, and 24
(bottom depths 1,632, 1,433, and 894 m).
gest continual existence of any lake now on
earth, Baikal has evolved a rich endemic
flora and fauna that have been the subjects
of scientific study for more than a century.
The past few decades have witnessed the
installation
of a large paper mill near the
southern end of the lake and the completion
of the Baikal-Amur
line of the trans-siberian railway along its northern reaches. The
expansion of such activities on its shores
and in the drainage basin of its tributaries
has heightened concern about preservation
of this unique ecological setting. With regard for this problem, a joint Soviet-American physical and geochemical study of Lake
Baikal was carried out in July 1988 under
the auspices of the Working Group 8 Bilateral Agreement for the Protection of the Environment with the support of the Institute
of Limnology at Irkutsk.
Fundamental to assessing its biochemical
state, a primary aim of this initial study was
to use oceanographic techniques to constrain the time scale of physical mixing in
the lake. Continuous conductivity,
temperature, and transmissivity profiles along with
discrete water samples for measurements of
dissolved 02, nutrients, fluorocarbon- 12, 3H,
3He, 4He, total inorganic C, 13C, and sus-
pended particulate material were obtained
for this purpose and will be presented elsewhere (see Weiss et al. 199 1). On the basis
of high oxygen concentrations throughout
the water column and other hydrologic observations, Lake Baikal has been presumed
to be dimictic with the entire water column
turning over twice annually as the sur&ce
waters reach the temperature of maximal
density (e.g. see Votintsev 1985). The tracer
studies showed that as a consequence of its
great depth, complete renewal takes - 8 yr.
This long renewal time has important implications for nutrient cycling within the lake
as discussed by Weiss et al. (199 1).
A second principal goal of this study was
to characterize the chemistry of the lake as
a baseline for examining future anthropogenic perturbations. In addition, the lake is
situated in a seismically active rift zone that
displays particularly high heat flows in its
northern region, including the northern end
of the lake. Observations of temperature
anomalies in the water column in this region
(Golubev 1978, 1984) and the nearby occurrence of numerous land-based, sulfidebearing, hot (40”-90°C) springs (Plyusnin et
al. 1979) have been used to infer the presence of hot springs in the lake. Thus a further intention was to determine whether hydrothermal activity affects the chemistry of
the lake. Toward these aims, water samples
for alkalinity and major and minor elements
were collected; the results are the subject of
this report.
Sample collection and storage
Sample bottles (500 ml, linear polyethylene) were precleaned by sequential 0.2 N
HCl and distilled, deionized water (DDW)
leachings overnight at 60°C, followed by
rinsing with DDW and drying in a laminar
flow bench. In July 1988 three profiles were
obtained at or near the deepest part of each
of the three major basins of the lake (Fig.
1) from the RV Vereshchagin with 5-liter
Niskin bottles, equipped with epoxy-coated
internal springs, neoprene (Buna-N) rubber
o-rings,
and reversing
thermom-eters,
mounted on a steel hydrowire. After fluorocarbon- 12, helium/tritium,
C02, and 0,
samples were taken, unfiltered water samples were drawn directly into the sample
Geochemistry of Lake Baikal
bottles, which were covered with clean plastic bags and shipped to the laboratory.
Upon their arrival in the laboratory, - 1
month after collection, aliquots of the samples were withdrawn for alkalinity and major ion analyses. Two months after collection, the samples were filtered for minor
element analyses through 0.4-pm Nuclepore filters that had been leached in 0.2 N
HCl into 0.2 N HCl-leached polyethylene
bottles. The samples were acidified 1 week
after filtration with 1.5 ml of 6 N HCl that
had been three times distilled in a Vycor
still.
Analytical methods
The analytical uncertainties reported here
are based on the relative standard deviation
of replicate analyses at the reported concentration levels. Titration alkalinities were
determined by a potentiometric
method
with an uncertainty of - 1% (Edmond 1970).
The major anions, Cl- and SOd2-, were
measured by ion chromatography to +5%.
The major cations Ca, Mg, Na, and K were
determined
by flame atomic absorption
spectrophotometry
(FAAS) with estimated
errors of 1, 1, 3, and 2%. K (+ 5%), Sr (-t 2%),
and Rb (+ 5%) concentrations for one of the
samples were determined by isotope-dilution, thermal-ionization
mass spectrometry
(ID-TIMS)
after separation by ion exchange. The isotopic composition of Sr was
also determined with the 87Sr : 86Sr ratio
normalized to an Eimer & Amend standard
value of 0.708 and the 86Sr: 88Sr ratio of
0.1194.
Minor elements were analyzed by inductively coupled plasma quadrupole
mass
spectrometry via a VG Plasmaquad with
conventional sample introduction
by peristaltic pumping through a concentric Meinhard nebulizer. Li was determined by isotope dilution with a 6Li spike provided by
Oak Ridge National Laboratory.
Values
were normalized to a gravimetric Li standard solution made from a Li,CO, salt presumed to have a natural isotopic composition of 7.42% 6Li, 92.58% 7Li. Li does
display significant natural isotopic variations so the data set may be somewhat offset from the true values, however, they are
415
internally
consistent within an analytical
uncertainty of 4%.
Ba and U were also determined by isotope
dilution with a 135Ba spike used for the
GEOSECS expedition and a 235U spike provided by Oak Ridge National Laboratory.
Values were normalized to calibrated gravimetric Ba and U standard solutions made
from high-purity BaC03 and u308
obtained
from SPEX Industries. Analytical
uncertainties are estimated to be 2 and 7% for Ba
and U. Sr and Rb concentrations were determined on unspiked samples with an uncertainity of -4%, using natural Ba as a
virtual isotopic spike and normalizing to a
gravimetric
monitor
solution containing
known amounts Ba, Sr, and Rb.
As a preliminary effort, a few samples were
also analyzed for trace (<nm01 kg- ‘) elements. Cs was determined by ion exchange
and ID-TIM&
Ge by hydride generation
and graphite furnace atomic absorption
spectrophotometry
(GFAAS, Andreae and
Froelich 198 1); Al by evaporative preconcentration and GFAAS; Cu, Ni, and Cd by
Preconcentration
of Na-bis(2-hydroxyethyl) dithiocarbamate
complexes onto
XAD-4 resin and GFAAS (Van Geen and
Boyle 1990); and Be by electron capture detection gas chromatography (Measures and
Edmond 19863). Analytical
uncertainties
are estimated to be - 10%.
Results
During the early phase of this work, it
became apparent that the major ions (Table
1) did not display significant variability
within the lake and so not all samples were
analyzed for every major species. Standard
deviations of the averages (Table 1) are in
fact comparable to analytical precisions. A
few data points fell significantly outside the
standard deviation of the average values and
were excluded from the standard deviation
estimates. Such values occurred randomly
in the large data set and hence are attributed
to sample handling or processing error rather than environmental
features. No significant species appears to be missing from the
major ion analyses since the charge balance
of individual
samples for which all species
were determined show no consistent offset
and fall well within analytical uncertainties.
416
Falkner et al.
Table 1. Major ion data (pm01 kg ‘).
--Sample
1
Depth
(m)
17
18
19
20
21
22
23
24
6
9
21
42
93
156
194
231
321
421
522
623
632
726
822
903
997
1,091
1,186
1,281
1,376
1,472
1,568
1,607
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
4
18
23
47
104
172
210
248
288
336
470
480
478
577
674
774
871
968
1,063
1,157
1,249
1,341
1,415
1,424
49
50
51
52
53
54
5
9
22
46
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Alk*
1,093
1,097
1,095
1,094
1,095
1,093
1,099
1,096
1,103
1,093
1,091
1,093
1,091
1,090
1,093
1,097
1,097
1,099
1,094
1,094
1,098
1,096
1,104
1,100
Ca
396.9
398.1
396.0
425.6-f
394.9
402.2
394.4
399.4
406.6
397.4
395.6
395.7
396.0
395.2
395.0
395.5
425.4f
396.6
406.7
399.1
395.6
400.2
396.3
396.5
Na
K
Cl
164.2
24.3
24.3
24.2
24.0
24.2
24.1
23.9
24.3
24.4
24.4
24.2
24.2
24.3
24.1
24.3
24.0
24.3
24.3
24.2
23.2
24.4
23.9
23.3
25.4
12.3
-
61.9.f
-
11.1
13.4
12.3
11.9
-
61.9t
-
Mg
Station 7
126.5
127.4
125.4
127.0
126.4
127.9
126.5
126.9
124.7t
126.3
124.4
125.6
-
152.5
156.5
148.0
150.4
154.2
156.5
13.4
12.5
SO,2
58.7
57.1
55.6
55.3
56.4
Chemimbal.
vd
-.
-.
-.
1.04
-.
0.01
-.
0.28
-
Station 17
1,089
1,099
1,102
1,105
1,097
1,091
1,091
1,102
1,096
1,094
1,099
1,093
1,094
1,093
1,102
1,100
1,090
1,092
1,089
1,092
1,078
1,086
1,103
1,104
389.0
390.7
395.2
395.5
394.8
408.9
399.3
396.4
405.9
405.0
405.0
419.9
410.7
405.3
414.7
450.3-F
403.3
409.9
449.6-f
401.8
405.6
402.6
124.6
124.6
123.7
-
461.2-F
125.8
126.9
126.7
126.2
125.4
125.3
126.0
126.1
129.3t
405.8
158.2
157.1
161.1
152.0
152.8
197.31.
159.7
25.7
24.1
23.5
24.9
24.0
24.2
23.3
23.5
24.0
25.8
24.7
24.2
23.9
23.5
24.8
22.4
23.6
23.6
24.2
23.2
28.5
25.7
23.4
22.9
12.1
12.5
-
22.3
25.9
24.7
21.6
23.3
21.6
12.5
12.5
-
-
19.9.f
13.0
11.2
12.5
56.4
56.4
59.4
57.0
0.22
--0.84
-2.32
--1.74
-_--
57.6
57.0
-
2.35
.0.85
-
56.4
59.4
57.6
Station 24
101
166
1,089
1,078
1,082
1,087
1,094
1,087
408.2
406.0
407.4
404.9
400.0
403.3
125.3
124.1
125.7
156.0
156.1
417
Geochemistry of Lake Baikal
Table 1. Continued.
Sample
Depth
55
56
57
58
59
60
61
63
64
65
204
241
279
327
375
518
612
767
836
887
Avg
SD
% dev.
* peg kg-l.
t Not included
(m)
Alk*
Ca
1,091
1,093
1,087
1,084
1,083
1,082
1,086
1,093
1,089
1,097
412.7
412.4
500.2.f
405.6
442.4-f
407.4
418.4.f
407.8
457.3-f
409.7
1,093
6
0.6
402
2
Na
Mg.
125.2
125.7 .
125.1
124.6
124.2
Cl
so,2-
-
-
-
152.0
24.1
24.3
24.1
23.9
23.5
24.1
23.8
24.1
23.9
12.5
11.7
11.7
58.2
58.2
59.4
1.75
155
4
3
24.1
1.0
4.1
12.3
0.7
5.8
57.4
1.3
2.3
1.2
0.9
-
150.1
192.4t
-
126.8
Summary
126
1
1
K
Chem.
imbal.
(%)
in avg and SD (see text).
The data permit precise determination
of a
weight-based salinity for Lake Baikal water
of 0.0963 -10.0006%, required for proper
application of the equation of state for freshwater (Chen and Miller0 1986).
Sr, Li, Rb, and U show no depth trends
outside of analytical uncertainties (Table 2).
Ba, however, appears to decrease gradually
with depth with near-bottom samples at stations 7 and 17 showing more abrupt depletions (Fig. 2). Trace-element results, presented in Table 3, are considered to be
preliminary
due to the delay between collection of the samples and their filtration
and acidification.
Although the values are
probably of the correct magnitude, detailed
profiles from samples filtered and acidified
upon collection will need to be obtained
along with hydrographic properties in a follow-up effort to better assess the sampling
procedure.
Discussion
Allowing for reasonable estimates of analytical uncertainties in older data, we find
that previously published major ion concentrations for the lake are for the most part
comparable to the ones presented here (Table 4). Historical Cl- and K values, however, generally exceed ours by at least a factor of two (Kozhov 1963; Votintsev 1961).
Independent determination
of K by ID-
TIMS confirmed our FAAS results. Previous data may have been compromised by
Na interference in the gravimetric
determination of K. Earlier Cl- values range
widely and hence are suspect. These problems as well as larger analytical uncertainties are probably the source of the greater
degree of charge imbalance displayed in earlier data sets.
As reported previously (Kozhov 1963),
the total mineral content of the lake is relatively low, with its major ion composition
dominated by Ca+2 and HC03-. Its rivers
and streams generally share a similar composition, although they carry higher levels
of Si, which is removed within the lake by
biological activity (Votintsev
1985). Although the geological terrain surrounding
the lake is quite complex, the chemistries
of its tributaries are thought to be controlled
predominantly
by the weathering of marble
and other carbonates among the metamorphic Archean and Proterozoic rocks found
in the mountains bordering the lake (Kozhov 1963; Votintsev 1985).
The lack of variability
in the major ion
composition of the lake would suggest that
the lake is well mixed with respect to the
residence times of the major ions in it. The
residence times of the major ions for which
inflowing river concentrations are available
were calculated with the averaged water
418
Falkner et al.
Table 2. Minor element data (nmol kg-l).
--Sample
Depth
(m)
Sr
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
6
9
21
42
93
156
194
231
321
421
522
623
632
726
822
903
997
1,091
1,186
1,281
1,376
1,472
1,568
1,607
1,410
1,340
1,350
1,360
1,360
1,360
1,310
1,400
1,430
1,320
1,340
1,310
1,380
1,300
1,370
1,370
1,370
1,420
1,410
1,370
1,370
1,370
1,390
1,350
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
4
18
23
47
104
172
210
248
288
336
470
480
478
577
674
774
871
968
1,063
1,157
1,249
1,341
1,415
1,424
1,370
1,350
1,340
1,300
1,370
1,340
1,310
1,560*
1,350
1,310
1,290
1,290
1,290
1,350
1,340
1,310
1,320
1,320
1,400
1,380
1,370
1,370
1,350
1,440
49
50
51
52
53
54
55
56
5
9
22
46
101
166
204
241
1,380
1,360
1,290
1,300
1,350
1,330
1,300
1,360
Li
Station 7
280
301
301
298
295
282
321
295
305
297
305
300
279
293
301
237
290
296
311
304
302
302
307
297
Station 17
291
290
299
305
290
297
301
289
294
295
286
296
297
301
305
291
293
295
305
295
307
299
294
295
Station 24
299
296
303
289
295
305
305
295
Ba
Rb
U
76.0
77.1
76.1
76.6
76.9
76.5
75.9
75.4
77.0
76.3
74.0
75.7
75.0
73.6
74.3
74.5
73.2
72.9
72.2
72.7
72.5
71.3
70.0
67.7
7.04
6.92
7.01
7.07
7.04
7.14
6.83
7.48
7.52
6.84
6.91
6.79
7.19
6.73
7.00
7.02
6.92
7.11
7.39
7.00
7.32
7.00
7.05
6.83
1.70
1.87
1.91
1.83
1.84
1.79
1.87
1.77
1.88
1.82
1.82
1.82
1.76
1.76
1.78
1.63
1.79
1.56
1.81
1.39
1.94
1.34
1.76
1.69
75.6
75.4
76.8
74.9
75.8
75.4
76.0
89.3*
76.5
72.4
73.9
74.8
74.2
75.5
74.5
74.6
73.4
74.2
73.5
73.6
73.4
73.7
73.1
67.2
7.09
7.09
7.11
6.79
7.17
7.15
6.88
8.17*
7.17
6.89
6.81
6.92
6.77
7.16
7.30
7.02
7.09
7.09
7.40
7.33
7.80
7.68
7.16
7.56
1.80
1.80
1.77
1.73
1.83
1.78
1.78
1.99*
1.86
1.73
1.73
1.62
1.70
1.75
1.67
1.44
1.77
1.65
1.62
1.58
1.89
1.85
1.73
1.77
75.8
74.9
74.7
74.2
75.4
75.3
74.6
74.7
7.16
7.15
6.70
6.79
7.07
7.08
6.84
7.08
1.92
1.87
1.71
1.83
1.87
1.77
1.86
1.79
--
419
Geochemistry of Lake Baikal
Table 2. Continued.
Denth
Samvle
Sr
(m)
279
327
375
518
612
767
836
887
57
58
ii:
61
63
64
65
1,350
30
2.8
Avg
SD
% dev.
* Not included
294
286
288
306
307
258
301
295
Summary
296
12
3.9
1,330
1,350
1,350
1,380
1,350
1,450
1,360
1,330
U
Rb
Ba
Li
75.3
75.7
75.4
75.4
75.0
73.9
73.5
75.3
7.21
7.00
7.17
7.19
7.11
7.52
7.07
6.82
1.83
1.64
1.83
1.87
1.88
1.87
1.78
1.69
74.7
2.6
3.6
7.10
0.23
3.3
1.77
0.12
6.9
in avg and SD (we text).
budget for the years 190 l-l 95 5 (Table 5)
(Afanyasev 1960). The results (Table 6) show
that with the exception of Cl-, the residence
time of the major ions is about 330 yr or
about the same as the residence time of water in the lake (Table 5). Thus to a first,
approximation
the fluxes of the major ions
must be dominated by riverine transport.
As Cl- would be expected to behave conservatively in a lacustrine environment, its
anomalously low residence time probably
reflects an overestimate of the Cl- concen-
tration for the tributaries, consistent with
the historical overestimates of Cl- in the
lake water. Given a mixing time of - 10 yr
(Weiss et al. 1991), the major ions undergo
on the order of 30 mixing cycles during their
residence in the lake, and so it is not surprising that their concentrations are quite
uniform.
Two years after our expedition,
clear,
warm water (16°C) seeps at 4 10 m at Frol&ha Bay in the northeastern end of the lake
were observed directly from a Pisces sub-
Table 3. Preliminary trace-element data (1 SD given in parentheses).
Sta.-Sample
7-l
7-3
7-5
7-10
7-12
7-13
7-15
7-19
7-20
7-22
7-23
7-24
17-25
17-26
17-31
17-48
24-49
24-60
24-63
Beim 1988$
Depth
(m)
6
21
93
421
623
632
822
1,186
1,281
1,472
1,568
1,607
4
18
210
1,424
5
518
767
Al
Zn
Cu
(nmol
Ni
Cd
kg-‘)
Be
(pmol
8.5(0.9)
11WI
2wYt
7.9(0.8)
3.0(0.3)
2.9(0.3)
3.3(0.3)
3.5(0.4)
2.7(0.3)
2.8(0.3)
2.6(0.3)
3.1(0.3)
74(7)
130(13)
290(29)-f
24(2)
7.4(0.7)
2.7(0.3)
2.8(0.3)
310(31)=f
7.0(0.7)
3.9(0.4)
3.1(0.3)
130(13)
55(5)
Ge*
CS
kg-l)
27(8)
23(2)
21m
14(3)
37(3)
140(1O)?
37(4)
13(3)
2W
2,000
3.1(0.3)
8.0(0.8)
2.6(0.3)
2.4(0.2)
1.3(0.1)
2.58(0.26)
93(18)
66(14)
44(4.5)-j-
3.0(0.3)
1.98(0.20)
22(4)
6
7
140
* Corresponding
Si concentrations
(pmol kg-‘): 7- 1 = 26.1; 7-22 = 66.3; 7-24 = 45.0.
t Likely to have been contaminated.
$ Averaged values for southern basin of Lake Baikal.
39(4)
‘Q(4)
2W)
48(8)
48(8)
Falkner et al.
420
Table 5. Water budget for Lake Baikal, 1901-1955
(after Afanasyev 1960).
Ba nmol kg- ’
60
65
70
75
80
85
Volume (km’)
Input (km3 yr-I)
Rivers
Underground
Rain
Total
Output (km3 yr-I)
outflow
Evaporation
Total
0
--L---Th=+--J
-lb0
200
400
600
800
1
-
+
U
q.i
1
1200
-
57.8
3.1
9.3
70.2
60.9
9.3
70.2
-J
4
1000
23,000
1
1600
q
cl
0
sta.7
l
Sta. 17
+
Sta. 24
--- _- - --.-
Fig. 2. Plot of Ba concentrations for all stations,
excluding sample 32 which is suspected to be in error.
mersible (K. Crane pers. comm.). The area
is at the foot of an east-west trending fault
that had extremely high heat flows. Unlike
sediment-hosted
oceanic hot springs, no
chimneylike structures were observed, which
suggests that these fluids do not carry large
amounts of reduced sulfur or other mmeralforming elements into the lake.
Although the chemistry of the warm fluids has yet to be directly characterized, hydrothermal discharge does not appear to be
significant enough to affect the budgets of
the major ions on the 300-yr time-scale of
water renewal by rivers. The one 87Sr : Vr
ratio we measured, 0.7085 1 +O.OOOlO (sta.
7,82 1 m), is lower than typical granitic continental values but falls within the range of
values representing the weathering terrain
surrounding the Baikal rift zone (Plyusnin
et al. 1979) and so does not necessarily in-
Table 4. Comparison with previous minor and major element data.
Vereshchagin*
1949
Major ions (pm01 kg-‘)
Alk
so,zCl
Ca
Mg
Na
K
Sum anions
Sum cations
% imbalance
Minor elements? (nmol kg--‘)
Li
Ba
* Cited by Kozhov
1963.
t I SD given in parentheses.
1,040
50
20
380
169
170
59
1,160
1,327
13
Votintsev
196 1
1,040
54
40
380
128
165
51
1,188
1,232
3.6
Beim
1988
210
250
160
20
1,100
260(20)
51
This work
1988
1,093
57.4
12.3
402
126
155
24.1
1,221
1,234
1.0
296(12)
74.7(2.6)
Geochemistry of Lake Baikal
Table 6. Major ion budget for Lake Bailkal.
Species
Alk
so42Cl
Ca
Mg
Na
K
Inventory*
(1 ,O 10 mol)
2,510
132
28
925
290
357
55
Ef?lucntst
(pm01 kg-‘)
1,300
70
51
500
180
5.11
Riverine
flux (I ,0 10
mol)
7.51
0.4
0.29
2.9
2i.51
Residence
time (yr)
334
330
97
320
279
353
* This work.
t Votintscv
1985.
$ (Na+K)
in mg kg I.
8 (Na+K)
in 1,010 g.
dicate hydrothermal activity. The atom ratio of Ge : Si in the few samples analyzed
(0.72-l. 1 x 10-6) is close to the continental
weathering ratio (0.3-l .2 X 10m6, Mortlock
and Froelich 1987) and much lower than
values observed for midocean hot springs
and basalts (27 X 10-6, Mortlock
and
Froelich 1986). If the warm springs contribute significantly to the budget of the minor elements, their signals are dissipated in
the water column such that no anomalies
in the vertical profiles are observed at our
northern basin station 24-closest
to the
warm-spring site.
The pH of the lake water is generally 7. l7.2, although photosynthetic activity results
in a larger range (7.1-8.6) in surface waters
(Kozhov 1963). Thermodynamic
calculations with the EQ3nr computer code (Wolery 1983) show that except occasionally,
when the pH exceeds 9 in surface waters,
the lake is everywhere undersaturated with
respect to calcite and aragonite. Consistent
with the thermodynamic
predictions, such
phases generally are not preserved in deep
lake sediments (Kozhov 1963). On the other
hand, the lake is also undersaturated with
respect to amorphous silica, but siliceous
organisms such as sponges and diatoms are
abundantly preserved in the sediments, on
average constituting -20% of the sediment
mass (Kozhov 1963). As occurs in the oceans
(DeMaster 198 l), high production rates and
kinetic inhibition must slow dissolution sufficiently to ensure preservation. Although
Lake Baikal is a net sink for silica and as a
result, probably also for the chemically
analogous, inorganic Ge (Froelich et al.
1985), it does not affect the major ions be-
421
cause they are not associated with biogenically precipitated silica to any appreciable
extent.
Our Li results agree with recently reported values, although previous Ba values are
-30% lower than our determinations
(Table 4). We could locate neither literature values for Sr, Rb, or U in Lake Baikal nor
information concerning the fluxes of the minor elements reported here. With the exception of Ba, the homogeneous distributions of the minor elements suggest that they
reside in the lake on the time scale of at least
several mixing cycles or a few decades. The
U concentrations (1.8 nmol kg-l) are about
an eighth of seawater concentrations
and
similar to levels in other oxic freshwaters
including surface waters of Lake Tanganyika (1.5 nmol kg-l), another intracontinental rift lake (K. Falkner and J. Edmond
unpubl.), and global river waters (M. Palmer and J. Edmond unpubl.).
Our preliminary
trace-metal results can
be compared to those reported by Beim
(1988) for dissolved metal concentrations
in southern Baikal waters in Table 3. Cu
and Ni concentrations exceed ours by a factor of 2-3. In our samples, these trace elements may have adsorbed onto the container walls between sample collection and
filtration, so our concentrations may represent lower limits. Zn and Al levels grossly
exceed our results and are undoubtedly attributable to contamination.
The Cu, Ni,
Cd, Zn, and Al concentrations we report are
similar to or lower than levels typically reported for oxic lakes (Sigg 1985; Sigg et al.
198 7) and comparable to oceanic ones (Bruland 1983; Hydes 1983; Measures and Edmond 1986a).
Of the major and minor elements studied,
only Ba displays any trend outside of analytical errors. The - 10 nmol kg-l Ba concentration decrease with depth is unlikely
to be an analytical artifact since the data
were replicated, with the samples run in random order and on three separate occasions.
The consistency between profiles would argue against a sampling artifact. Moreover,
it is difficult to imagine one that would have
affected only Ba and no other minor element. The ratio of Sr to Ba (- 18) in the
lake is almost a factor of two greater than
422
Falkner et al.
the high end of the range observed for a wide
variety of world rivers (0.5-10; M. Palmer
and J. Edmond unpubl.), which suggests that
the process redistributing
Ba in the water
column results in its net removal in the lake.
At stations 7 and 17, Ba concentrations
decrease abruptly near the bottom. Light
transmission also decreased sharply near the
bottom at these sites. The concentration of
particles in these bottom samples is estimated to be 100 pg kg- 1 or roughly double
ambient particle concentrations (W. Gardner pers. comm.). Mixing processes in Lake
Baikal resulted in bottom waters at stations
7 and 17 having surfacelike fluorocarbon12, silica, and O2 signatures at the time of
sampling (Weiss et al. 199 1). In contrast,
the bottom sample for station 24 does not
show diminished
Ba concentrations,
reduced light transmission, or surfacelike hydrographic properties. Station 24 was not
located in the deepest part of the basin, so
perhaps the newest and coldest bottom water (containing resuspended bottom sediments), observed to collect in the deepest
part of the other basins, was not sampled.
It remains to examine possible processes
governing the observed Ba distributions.
In the oceans, barite (BaSO& is removed
from surface waters in association with biological productivity
(Dehairs et al. 1980).
The exact mechanism for barite production
in undersaturated
seawater (Church and
Wolgemuth 1972) is not well understood,
although it occurs in microenvironments
provided in aggregates of decaying organic
matter and diatom tests (Bishop 1988). It
seems unlikely that such a process is responsible for the Ba distributions
in Lake
Baikal because the highest concentrations
of Ba occur in the surface waters of the lake.
One could hypothesize that the biological
Ba removal rate seasonally overcomes the
riverine flux to surface waters and that the
sharp decreases in Ba in the bottom samples
at stations 7 and 17 are remnant signatures
of convected surface water that had been
temporarily depleted of Ba. For surface-water Ba concentrations to be replenished between the time of removal and sampling
would require seasonally elevated riverine
Ba inputs. Hydrographic properties indicate
a mixed layer of about 250 m for the lake
(Weiss et al. 199 I), however, equivalent 1.0
a volume of 5,750 km3 and thus a riverine
renewal time of about 100 yr. Riverine Ba
concentrations would have to vary at least
1O-fold on a seasonal basis to restore mixedlayer Ba depletions- which seems highly
unlikely since the other major cations do
not do so (Votintsev 1985). Furthermore,
the elevated riverine inputs would have to
be mixed nearly instantaneously
throughout the lake mixed layer, which is unrealistic.
Alternatively,
there are -several candidates for Ba removal at depth that could
explain the water-column distributions.
Biological precipitation
of barite by freshwater protozoa of the genus Loxodes has been
documented in other lakes (Finlay et al.
1983; McGrath et al. 1989). Barite may be
precipitated
inorganically
in sulfate-rich
microenvironments
provided by decaying
organic matter at the sediment surface. It
has also been suggested that Ba can be controlled in freshwater environments by adsorption onto Mn or Fe oxide surfaces
(Sholkovitz and Copland 1982) that precipitate at the sediment-water interface in Lake
Baikal (Leibovich-Granina
1987). Resuspension of such phases at stations 7 and 17
might explain the low Ba concentrations in
the bottom waters at these sites.
If the Ba distributions are at steady state,
using a deep-water residence time of 8 yr
(Weiss et al. 199 1) and a mean deep-water
Ba depletion of 2% yields a Ba residence
time with respect to removal in deep waters
of -400 yr. This value is comparable to the
residence time of water in the lake, implying
that the flux of Ba to the sediments is about
half of its riverine input. While it would
seem that uptake of Ba at the sediments of
Lake Baikal plays an important role in Ba
cycling, it is not at all clear how any of the
proposed removal mechanisms provide a
net sink for Ba. In the case of barite production, diagenetic consumption of the low
concentration of dissolved sulfate in pore
waters would result in the dissolution of
barite and the return of Ba to the water column. Likewise as Mn and Fe oxides are
buried, they become reduced and release
adsorbed elements. The existence of a significant cap of oxidized sediments could re-
Geochemistry of Lake Baikal
sult in retrapping of this dissolved Ba below
the sediment-water interface, thus preventing its escape into the overlying water column. The question of what controls Ba in
Lake Baikal will only be resolved through
further study of its distributions
and particulate fluxes along with sediment pore-water and solid phase fluxes.
References
AFANASYEV, A. N.
1960. The water budget of Lake
Baikal. Tr. Baik. Limnol. Sta. Akad. Nauk SSSR
Vast.-Sib. Fil. 18: 155-241.
ANDREAE, M. O., AND P. N. FROELICH. 1981. Determination of germanium in natural waters by
graphite atomic absorption spectrophotometry with
hydride generation. Anal. Chem. 53: 287-29 1.
BEIM, A. M. 1988. The mineral composition of purified wastewaters from sulfate-cellulose industry,
p. 43-54. In 0. M. Kozhova and L. Y. Yashchepkova [eds.], Long-term prognosis for the state
of the ecosystem. Nauka.
BISHOP, J. K. B. 1988. The barite-opal-organic carbon
association in oceanic particulate matter. Nature
332: 341-343.
BRULAND, K. W. 1983. Trace elements in sea-water,
p. 157-220. In J. P. Riley and R. Chester [eds.],
Chemical oceanography. V. 8. Academic.
CHEN, C. T., AND F. J. MILLERO. 1986. Precise thermodynamic properties for natural waters covering
the limnological range. Limnol. Oceanogr. 31: 657662.
CHURCH, T. M., AND K. WOLGEMUTH. 1972. Marine
barite saturation. Earth Planet. Sci. Lett. 15: 3544.
DEHAIRS, F., R. CHESSELET, AND J. JEDWAB. 1980.
Discrete suspended particles of barite and the barium cycle in the open ocean. Earth Planet. Sci.
Lett. 49: 528-550.
DEMASTER, D. J. 198 1. The supply and accumulation
of silica in the marine environment. Gcochim,
Cosmochim. Acta 45: 17 15-l 732.
EDMOND, J. M. 1970. High precision determination
of titration alkalinity and total carbon dioxide content of seawater by potentiometric titration. DeepSea Res. 17: 737-750.
FINLAY, B. J., N. B. HETHERINGTON, AND W. DAVISON.
1983. Active biological participation in lacustrine
barium chemistry. Geochim. Cosmochim. Acta
47: 1325-l 329.
FROELICH, P. N., G. A. HAMBRICK, M. 0. ANDREAE,
AND R. A. MORTLOCK. 1985. The geochemistry
of inorganic germanium in natural waters. J. Geophys. Res. 90: 1133-1141.
GOLUBEV, V. A. 1978. Vertical temperature gradients
and static stability of the waters of Lake Baikal.
Dokl. Akad. Nauk SSSR Geofiz. 239: 1309-l 3 12.
-.
1984. The discovery of underwater hydrothermal signals in Baikal by the method of con-
423
tinuous thermal profiling. Izv. Akad. Nauk SSSR
Fiz. Zemlii 19: 104-107.
HYDES, D. J. 1983. Distribution of aluminium in
waters of the north east Atlantic 25”N to 35”N.
Geochim. Cosmochim. Acta 47: 967-973.
KOUIOV, M. 1963. Lake Baikal and its life. Monogr.
Biol. V. 11. Junk.
LEIBOVICH-GRANINA, L. Z. 1987. Iron and manganese
in Lake Baikal. Water Resour. (Engl. Transl. Vodnye Resursy) 14: 259-264.
MCGRATH, M., W. DAVISON, AND J. HAMILTONTAYLOR. 1989. Biogeochemistry of barium and
strontium in a softwater lake. Sci. Total Environ.
87/88: 287-295.
MEASURES, C. I., AND J. M. EDMOND. 1986a. Aluminium in the northwest Atlantic. Geochim. Cosmochim. Acta 50: 1423-1429.
-,
AND -.
19863. Determination of beryllium in natural waters in real time using electron
capture detection gas chromatography. Anal.
Chem. 58: 2065-2069.
MORTLOCK. R. A., AND P. N. FROELICH. 1986. Hydrothermal germanium over the southern East Pacific Rise. Science 231: 43-45.
-,
AND -.
1987. Continental weathering
of germanium: Ge/Si in the global river discharge.
Geochim. Cosmochim. Acta 51: 2075-2082.
PLYUSNIN, G. S., G. P. SANIMIROVA, Y. A. PAKHOL’CHENKO, I. S. LOMONOSOV, AND Y. P. RZHECHITSKIIY. 1979. Origin of recent hydrothermal
solutions of the Baykal Rift Zone according to
isotope ratios. Geochem. Int. 16: 83-90.
SHOLKOVITZ, E. R., AND D. COPLAND. 1982. The
chemistry of suspended matter in Esthwaite Water, a biologically productive lake with a seasonally
anoxic hypolimnion. Geochim. Cosmochim. Acta
46: 393-410.
SIGG, L. 1985. Metal transfer mechanisms in lakes;
the role of settling particles, p. 283-310. In W.
Stumm [ed.], Chemical processes in lakes. WileyInterscience.
-,
M. STURM, AND D. KISTLER. 1987. Vertical
transport of heavy metals by settling particles in
Lake Zurich. Limnol. Oceanogr. 32: 112-l 30.
VAN GEEN, A., AND E. BOYLE. 1990. Automated preconcentration of trace metals from seawater and
freshwater. Anal. Chem. 62: 1705-l 709.
VOTINTSEV, K. K. 196 1. The hydrochemistry of Lake
Baikal. Tr. Baik. Limnol. Sta. Akad. Nauk SSSR
Vast.-Sib. Fil. 20: 1-312.
-.
1985. Main features of the hydrochemistry
of Lake Baikal. Water Resour. (Engl. Transl. Vodnye Resursy) 12: 106-l 16.
WEISS, R. F., E. C. CARMACK, AND V. M. KOROPALOV.
1991. Deep-water renewal and biological production in Lake Baikal. Nature 349: 665-669.
WOLERY, T. J. 1983. EQ3NR, a computer program
for geochemical aqueous speciation-solubility calculations, user’s guide and documentation UCRL534 14. Lawrence Livermore Lab.
Submitted: 13 November 1990
Accepted: 7 January 1991
Revised: 19 March 1991